General model by lkhl · 1.96B parameters
VideoLLaMA3 2B Image HF fits comfortably in your 24 GB Mac using Q8_0 quantization, using 11% of your RAM.
~82 tok/s
Q8_0 · 2.7 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 2.7 GB | ~82 tok/s | ✓ |
| Q6_K | 2.2 GB | ~108 tok/s | ✓ |
| Q5_K_M | 2.0 GB | ~127 tok/s | ✓ |
| Q4_K_M | 1.8 GB | ~149 tok/s | ✓ |
| Q3_K_M | 1.6 GB | ~179 tok/s | ✓ |
| Q2_K | 1.4 GB | ~233 tok/s | ✓ |
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free